A Study on Data Mining Method for Measuring Sludge Concentration in Water-soluble Metal Working Fluid

抄録

A condition monitoring method was designed to measure the contamination in water-soluble coolant from the time-series sensing data of spindle power consumption. The fluctuation pattern of time-series data streams can be visualized as a white and black pattern by utilizing the spindle power change rate average. The designed data mining method for spindle power consumption change can understandably display the fluctuation pattern of trend component in time-series spindle power data from the grinding operation. The designed method was applied to measure the sludge concentration in a water-soluble grinding fluid. The white or black pattern obtained from the data processing had a strong relation with the sludge concentration in water-soluble grinding fluid. Namely, the data mining scheme designed in this study could make possible to predict the sludge content in grinding fluid from an easy-handling grinding test.